The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses local git commands to gather context about the project's recent history and current state.
Evidence: Executes git status and git log -10 --oneline to identify active changes and recent commits.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection (Category 8) because it ingests untrusted content from the repository being analyzed.
Ingestion points: Reads content from CLAUDE.md, README.md, package.json, .plans/, .tasks/, TODO files, and git logs.
Boundary markers: Does not use specific delimiters or instructions to ignore embedded commands when processing the gathered file content.
Capability inventory: Capable of executing system commands (git), accessing the local file system, and spawning sub-agents via the Task tool.
Sanitization: No sanitization or validation of the ingested repository content is performed before it is synthesized into summaries or passed to the sub-agent Oracle.

ctx